Sexual expression: its emotional context in heterosexual, gay, and lesbian couples.

Sexual expression often occurs in the context of close relationships. An understanding of sexuality is facilitated by the inclusion of relationship dynamics, including daily feelings experienced toward one's partner. It is therefore important to understand the proximal relationship context of the sexual experience. Researchers have conceptualized close relationships as dynamic systems in which participants mutually influence each other at behavioral, cognitive, and affective levels (Dindia, 2000; Huston, 2000; Rusbult, 1983). For example, at one point in time, sexual initiation may flow from thoughts of sexual yearning in the absence of negative feelings, whereas at other times anger or other negative feelings may lead to increased sexual activity with one's partner (Barlow, 2002). There is a dynamic interplay between emotions and sexual expression, particularly emotions created in part by the partner.

The purpose of the current study is to explore negative feelings and sexual expression in the context of close relationships. Specifically, we focus on how negative feelings in general, and several specific types of negative feelings (anxiety, anger, and sadness), are linked to the sexual experiences of individuals in close relationships. Additionally, we explore similarities and differences between heterosexual and same-sex couples.

Positive and Negative Feelings and Sexual Expression

In close relationships, there is ebb and flow in the way that partners relate sexually and in the range and intensity of emotions experienced (Ridley et al., 2006). One day individuals want and seek sexual involvement; the next day sexual thoughts and behavior are absent. Feelings toward one's partner may also reflect this dynamic quality (Fortenberry et al., 2005). Kelley (1979) argues that a defining feature of dynamic systems such as close relationships is interdependence--partners mutually influence each other's thoughts, feelings, and behaviors. It might be expected, however, that not all feelings are equally influential or subject to being influenced by sexual wants and behavior. Some feelings produce an emotional breeze so strong that sexual desire is ignited and behaviors push toward sexual union. The breeze at other times is so mild that sexual desire goes unnoticed and sexual interaction is not attempted. In what ways might negative feelings play a unique role in sexual experiences?

Feelings differ not only in intensity but also in valence (Russell, 1980). As participants in a relational system move through time together, a mixture of positive and negative emotions occur and are in flux. This mix occurs both within individuals and between relational partners. Giles (2004) defined the sexual experience as two interrelated components: (a) thoughts about wanting sexual interaction with preferred gender others (sexual desire), and (b) the specific sexual behaviors that occur. Although these two components can, and likely do, mutually influence each other, one does not necessarily produce the other. Positive and negative feelings have been found to be associated with these two components in complex ways (Burleson, Trevathan, & Todd, 2007).

The study of romantic relationships has linked positive affect to sexual desire and behavior (Berscheid, 1988; Regan, 2000). It is not surprising that when partners indicate love and positivity toward each other, and define the relationship in positive terms (e.g., satisfied, intimate, and committed), they are more open to sexual experiences and more likely to report sexual involvement. As Giles (2004) states, "Love is but one more way of wanting what sexual desire wants: mutual vulnerability and care" (p. 171). Cooper, Shapiro, and Powers (1998) showed that after people engaged in sexual interaction with their partners, they expressed more positive feelings and felt more connected to them. It would seem that positive feelings toward one's partner anchor the sexual experience, pull partners toward each other, and create a relational context for pleasurable sexual interaction.

Although the direction of the association between positive feelings and sexual experiences is clearly documented, it is not as clear how negative feelings and sexual expression are linked. Negative feelings can be conceptualized as a defense against a real or perceived threat or danger (Lang, 1994). When in a state of being alert to danger or threat, it may be difficult for people to appreciate the potential pleasures of a sexual encounter. Empirical studies have found that most people experience a decline in sexual interest and behavior in negative mood states (Araujo, Mohr, & McKinlay, 2004; Beck, 1967; Burleson et al., 2007; Cassidy, Flanagan, Spellman, & Cohen, 1957; Fortenberry et al., 2005; Kennedy, Dickens, Eisfeld, & Bagby, 1999; Ridley et al., 2006), but there is support for a positive association in a minority of individuals (Bancroft, Janssen, Strong, & Vukadinovic, 2003).

Only a few studies have explored daily fluctuations in the relationship between negative feelings and sexual expression (Burleson et al., 2007; Fortenberry et al., 2005), and these studies focused either on midlife women or youth. Thus, the first goal of the present study is to examine the association between negative feelings and sexual expression (i.e., sexual behavior, sexual desire, wanted sexual behavior, arousal, and lust) in heterosexual and same-sex relationships while controlling for positive feelings.

Types of Negative Feelings and Sexual Expression

Anxiety, anger, and sadness are all negative feelings in the sense that they are commonly considered undesirable emotional states, and when these feelings are experienced, effort is often made to reduce or eliminate the intensity of them (Coyne, 1990). Each of these emotions, however, is unique in that they represent different internal experiences and potentially relate to sexual experiences in different ways.

Anxiety in close relationships is an emotional response to a perceived threat to the individual, the relationship, or both (Barlow, 2002). Sexual responding is limited because the person experiencing the anxiety can become preoccupied with searching for the causes and contributions to the anxiety, and thus is distracted from sexual expression (Barlow, 2002). Laboratory studies on anxiety and sexual arousal show, however, that anxiety can inhibit sexual arousal, have no effect on sexual arousal, or for some individuals enhance sexual arousal (Barlow, 2002; Barlow, Sakheim, & Beck, 1983; Beggs, Calhoun, & Wolchik, 1987; Elliot & O'Donohue, 1997; Lykins, Janssen, & Graham, 2006; Palace & Gorzalka, 1990).

When anger is experienced in the context of a close relationship, it may be the result of a perceived deliberate attempt by a partner to prevent the other partner from getting what he or she wants--it blocks a desired behavioral occurrence (Huston, 2000). Russell and Carroll (1999) state that although people who are angry feel unpleasantly aroused, they often feel in control of the situation and dominant. The likely action associated with anger is overt aggression or passive aggression in the form of withdrawal, both in service of an attempt to achieve a desired behavioral outcome. When angry, some may identify sex as an acceptable way to express their anger, reduce its intensity, and achieve a desired outcome. The literature on domestic violence provides support for this contention in that sex is often used by abused partners to reduce the danger of uncontrolled rage and to restore a more positive relational context (Christopher & Kisler, 2004). For couples that are less severely distressed, sex and anger may be incompatible in that on days that anger is intense, sex is less likely to occur. As an appetitive behavior, sex can be an act of pleasure and an indication of positive regard for one's partner (Lang, 1994). In contrast, anger is an adaptive behavior that occurs in response to threat. Thus, sex and anger might be experienced as internally inconsistent and even oppositional.

The clinical literature on sadness, a common component of depression, identifies a likely response pattern to sadness that includes the following: (a) engaging in an internal, self-critical search for the causes of the sadness; (b) seeking contact, validation, and reassurance from one's partner; and (c) returning to deeper sadness with a continued failure to feel adequately validated by the partner (Coyne, 1990). Sex may play a different role depending on which of these response patterns is entered. Sex may be used in seeking contact, validation, and reassurance from one's partner. If reassurance seeking fails and the causal search intensifies, however, sadness deepens and sex will likely become less salient. Even when sexual activity occurs while one is sad, sexual desire should be low because sadness tends to suppress physiological arousal and with it sexual desire (Wenzel, Jackson, & Brendle, 2004). In long-term relationships with repeated failures at receiving reassurance from one's partner, it may be difficult for individuals to continue to act in a sexual way and certainly difficult to desire sex. A minority of individuals (e.g., gay and heterosexual men), however, report increased sexual desire and sexual risk taking when depressed or anxious (Bancroft, Janssen, Strong, & Vukadinovic, 2003; Bancroft & Vukadinovic, 2004). Given the lack of clarity regarding the associations between anger, anxiety, and sadness and sexual expression, the second goal of the study is to investigate how the specific types of negative feelings (i.e., anxiety, anger, and sadness) relate to sexual expression.

Negative Feelings and Sexual Expression in Same-Sex Couples

The last goal of the current research is to explore feelings and sexual expression within different types of close relationships. In large part, the empirical research on the ebb and flow of affect and sexual expression in close relationships has been with heterosexual couples that are married (Ridley et al., 2006) or romantic couples in early stages of development (Regan & Berscheid, 1999). There is, however, some empirical evidence that men in same-sex relationships differ from women in same-sex relationships as well as in heterosexual relationships in sexual behavior (Blumstein & Schwartz, 1983).

Based on the empirical work to date on sexual behavior in same-sex couples, males in same-sex relationships may view sex as very salient in their life--they desire it and frequently engage in it (Peplau, Fingerhut, & Beals, 2004). Sex may therefore have the increased salience to impact the emotional system. Past research on lesbian relationships, however, has shown that sex is not as indicative of relationship success as other relational indicators (Peplau et al., 2004). This finding with lesbian women then would suggest that negative feelings may neither enhance nor suppress sexual expression, and sexual expression may not have the emotional potency to soothe negative feelings. Thus, on days that negative feelings are present, whatever happens sexually may have little to do with these feelings.

Additionally, both emotional and sexual expression are often different for men and women. Women are expected to experience and express more emotions than men, but men are expected to be more sexually active and initiate sex more often than women (Plant, Hyde, Keltner, & Devine, 2000). Women may be more forthcoming in expressing negative feelings, but to the extent that sexual drive is less urgent, the negative feelings may not influence the sexual domain. Men may place greater urgency on sexual expression but be reluctant to express negative feelings associated with the sexual experience (Vohs, Catanese, & Baumeister, 2004). Given the dearth of studies examining the association between negative feelings and sexual expression in same-sex couples, similarities and differences as a function of couple type and gender will be explored in the context of the two previous research goals.

Method

Sample

Two samples were collected for the purposes of the study. The first sample included 81 heterosexual couples (N = 162). Couples were recruited through posted announcements on and around a university campus in the southwestern United States. The majority of this sample was in their early twenties (M = 20.67, SD = 2.46), and Caucasian (75%). Most couples (95%) self-identified as seriously dating and had been in their current relationships for approximately 18 months (M = 17.63, SD = 12.14). This sample was highly educated, with more than 75% of participants having at least some college education.

The second sample included 106 same-sex couples (N = 214), of which 63 were lesbian couples and 43 were gay male couples. Couples were recruited via e-mail through a variety of organizations nationwide that represent gay, lesbian, and bisexual individuals (e.g., support groups, campus clubs, electronic list service, online newsletters, and social groups). The average relationship length for gay males was approximately 6.5 years (M = 6.48, SD = 9.67), and most men (95%) indicated that their relationships were serious. They were predominantly Caucasian (86%), and most had completed at least some college (86%). Gay male participants were approximately 35 years old on average (M = 35.55, SD = 13.45). Women in lesbian couples had been together for approximately 4 years (M = 4.15, SD = 5.69), and most indicated that their relationships were serious (97%). They were also predominantly Caucasian (78%), and most had completed at least some college (93%). Lesbian participants were approximately 31 years old (M = 31.31, SD = 11.15). For the purposes of the current study, the heterosexual and same-sex samples were combined, resulting in a total sample of 187 couples (N = 374).

Procedure

All data were collected via an Internet-based system. Prior to starting data collection, Internet materials and procedures were piloted with a small group of volunteers to ensure accuracy and ease of use. To ensure anonymity, participants entered unique ID numbers and passwords given to them by the researchers. When participants first logged onto the system, they completed an initial questionnaire consisting of demographic information and several personal and relationship scales. When the initial questionnaire was completed, the participants electronically submitted the data, which became inaccessible to them. Participants then were instructed to log onto the Web site at the same time each day for 14 consecutive days to complete a daily measures questionnaire. The daily questionnaires were the same for all 14 days and included a series of emotion and behavioral measures designed to capture the relational events that took place within the last 24 hours. Participants were instructed to complete the entire survey separately from their partner. Electronic time and date stamps were used to monitor and verify compliance with daily questionnaire completion and the independence of partner submissions. A questionnaire was considered valid if it was submitted before midnight on the day subsequent to the previous submission (i.e., only one submission per day was considered valid). On days when multiple submissions occurred, the questionnaire with the most complete set of data was used. All other submissions were not used. Individuals completed an average of 12.5 days and 77% completed all 14 days.

Measures

Negative and positive feelings. All participants were asked to respond to how they felt toward their partner in the last 24 hours on a list of feelings that included anger, anxiety, sadness, love, happiness, affection, intimacy, and satisfaction. Responses were given on a 5point, Likert-type scale (1 = not at all, 3 = moderately, 5 = very strongly). To create the scales, all items were subjected to a principal components analysis with varimax rotation. Selection of items was based upon the rotated component matrices. The components that were retained had eigenvalues greater than 1.0 and items loadings of .50 or higher. The results of the analysis revealed two factors: Negative Feelings and Positive Feelings. The first factor, Negative Feelings, included three items: anger, anxiety, and sadness ([alpha] = .77). The second factor, Positive Feelings, included five items: love, happiness, affection, intimacy, and satisfaction ([alpha] = .90). Scale scores for each participant were created by taking the mean of the scale items.

Sexual expression. To measure sexual expression, three components were examined: sexual thoughts, wanted sexual behaviors, and occurrence of sexual behaviors.

To measure sexual thoughts, participants indicated the intensity of their thoughts about sexual arousal, sexual desire, and lust as experienced toward their partner in the last 24 hours. Responses were given on a 5-point, Likert-type scale (1 = not at all, 3 = moderately, 5 = very

strongly). Each of these variables was used individually as a dependent variable.

To measure wanted sexual behavior, participants indicated their desired level of sexual involvement with their partner in the last 24 hours. Participants responded daily to whether they desired each of the following: hugging/cuddling, kissing, sexual touching, and sexual penetration (sexual intercourse, oral sex, or anal sex). A scale was created using the sum of these items to indicate the number of daily sexual behaviors desired.

To measure the occurrence of sexual behaviors, participants responded daily to whether they had engaged in each of the following sexual behaviors with their partner: hugging/cuddling, kissing, sexual touching, and sexual penetration (sexual intercourse, oral sex, or anal sex). Behaviors were summed and the total number of behaviors was used as a sexual behavior score. (1)

Results

All analyses were done using hierarchical linear modeling (HLM) software to account for the nonindependence of repeated measures over days and for potential similarity due to individuals being part of a couple (Raudenbush, Bryk, Cheong, & Congdon, 2001). All analyses were three-level models with days (Level 1), nested within individuals (Level 2), nested within couples (Level 3).

We used a three-level model in HLM to estimate the effects of daily feelings on sexual expression and to examine the effects of gender and relationship type on this association. Models were estimated separately for each of the sexual variables. For each model, we entered negative feelings and positive feelings as the Level 1 predictors. By entering both negative and positive feelings at Level 1, we were able to estimate the unique effect of negative feelings while controlling for the effects of positive feelings. Gender was entered at Level 2 to enable comparisons between men and women, which resulted in a main effect estimate as well as an interaction between each Level 1 predictor and gender (e.g., Negative Feelings X Gender). To test differences between same-sex and heterosexual couples, we entered relationship type as a predictor at Level 3, which resulted in a main effect estimate, estimates of the interactions between relationship type and each of the other predictors, and the 2 three-way interactions formed by pairing relationship type with each of the Level 1 predictors and gender. Age was controlled for at Level 2. Random intercepts were included for each of the models, and zchi-square statistics were used to decide whether to include random slopes (Raudenbush & Bryk, 2002). To interpret significant interactions, the estimated effect of negative or positive feelings on change in sexual expression was plotted for the two values of gender and couple type, and simple slope estimates were calculated (Aiken & West, 1991). Only the significant three-way interactions containing either negative feelings or its components are presented in the current article due to space constraints. All other graphs are available from the first author.

To test the unique effects of each of the components of negative feelings, the models above were modified to include each of the daily negative feeling components (anger, anxiety, and sadness) separately. All other predictors and controls remained the same. Separate models were run for each of the sexual variables.

Daily Negative Feelings and Sexual Expression

Table 1 presents the means and standard deviations for each of the study variables. On average, participants reported relatively high levels of daily positive feelings and each of the sexual expression variables. Specifically, participants reported approximately two to three sexual behaviors per day, or 14 per week (M = 14.22, SD = 6.39). Relatively low levels of daily negative feelings, anger, anxiety, and sadness were reported on average, although participants' scores spanned the whole scale (see Table 1).

The unique daily associations of negative and positive feelings and each of the sexual variables were tested. Negative feelings were positively associated with sexual arousal and lust, such that on days when negative feelings were high, reports of sexual arousal and lust were also high. Negative feelings were not significantly associated with sexual desire, wanted sexual behavior, or sexual behavior. Positive feelings had a strong positive association with all of the components of sexual expression, such that on days when positive feelings were high, each component of sexual expression was also high (see Table 2).

The interaction of negative feelings and gender was not significant for any of the sexual variables, nor was the interaction of negative feelings and couple type. The three-way interaction of Negative Feelings X Gender X Couple Type, however, was significantly associated with sexual arousal and sexual behavior (see Table 2). Negative feelings had a strong positive association with sexual arousal for women in heterosexual relationships ([beta] = .23, p < .05) and men in same-sex relationships ([beta] = .16, p < .05). The slopes for women in same-sex relationships ([beta] =.06, n.s.) and men in heterosexual relationships ([beta] = .06, n.s.), however, were non-significant (see Figure 1). For sexual behavior, we again found that the association was significant for women in heterosexual relationships ([beta] = .40, p < .05). This time, however, men in both heterosexual ([beta] = .23, p < .05) and same-sex relationships ([beta] = .16, p < .05) had significant positive slopes, whereas women in same-sex relationships ([beta] = .00, n.s.) had no association between negative feelings and sexual behavior (see Figure 2).

[FIGURE 1 OMITTED]

Anger, Anxiety, Sadness, and Sexual Expression

The results for anger showed a unique positive association with sexual desire only. Anger was not associated with any of the other sexual variables. Both anxiety and sadness were uniquely associated with wanted sexual behavior and sexual behavior, although these associations were positive for anxiety and negative for sadness. Thus, on days when anxiety was high, wanted sexual behavior and sexual behavior were also high, while the opposite pattern was true for sadness. On days when sadness was high, these sexual variables were relatively low (see Table 3).

[FIGURE 2 OMITTED]

The unique effects of anger, anxiety, and sadness as moderated by gender were examined. Only sadness demonstrated a statistically significant interaction with gender, and this effect was for wanted sexual behavior (see Table 3). The positive association between sadness and wanted sexual behavior was significant for men ([beta] = .10, p < .05), but not for women ([beta] = .04, n.s.).

The unique effects of anger, anxiety, and sadness as moderated by couple type were also examined. Anxiety and sadness both interacted with couple type in predicting wanted sexual behavior, but they did not predict any other sexual variable (see Table 3). In examining the interaction of anxiety and couple type, the negative association of anxiety and wanted sexual behavior was significant for heterosexual couples ([beta] = -.30, p < .05), but not for same-sex couples ([beta] = -.04, n.s.). The association between sadness and wanted sexual behavior was similar in that heterosexual couples had a significant effect ([beta] = .29, p < .05), whereas same-sex couples did not ([beta] = -.04, n.s.).

The three-way interactions of Anger X Gender X Couple Type and Anxiety X Gender X Couple Type yielded no statistically significant interactions for any of the sexual variables. Sadness X Gender X Couple Type, however, was significant when predicting sexual arousal, lust, sexual desire, and sexual behavior (see Table 3). For sexual arousal, sadness had a significantly positive effect for heterosexual men ([beta] = .05, p < .05) and heterosexual women ([beta] = .24, p < .05), a significantly negative effect for same-sex women ([beta] = -.10, p < .05), and no effect for same-sex men ([beta] = -.01, n.s.; see Figure 3). For lust, sadness had a significantly positive effect for same-sex men ([beta] = .07, p < .05) and heterosexual women ([beta] = .14, p < .05), a significantly negative effect for same-sex women ([beta] = -.10, p < .05), and no effect for heterosexual men ([beta] = .03, n.s.). Sadness was not associated with sexual desire for heterosexual men ([beta] = .06, n.s.), same-sex women ([beta] = -.03, n.s.), or same-sex men ([beta] = .02, n.s.), but it was associated for heterosexual women ([beta] = .21, p < .05; plots for lust and sexual desire are available from the first author). Sadness was significantly associated with sexual behavior for heterosexual men ([beta] = .34, p < .05), heterosexual women ([beta] = .47, p < .05), and same-sex men ([beta] = .07, p < .05), but not for same-sex women ([beta] = -.01, n.s.; see Figure 4).

Discussion

The current study focused on the association between negative feelings toward one's partner and sexual expression in relationships. Consistent with prior research and theory (Regan, 2000; Ridley et al., 2006), we found that sexual thoughts and behaviors are strongly related to positive feelings. The daily associations of positive feelings and sexuality were found with sexual arousal, sexual desire, and sexual behavior for both males and females in heterosexual relationships and same-sex relationships. Thus, in all types of close relationships there is something robust about wanting and having sex when feelings about one's partner are positive.

Concerning the first goal of the study, we were not entirely surprised to find that negative feelings had a unique daily association with sexual experiences. We were surprised, however, that the associations were positive, not negative (i.e., on days that negative feelings toward the partner were high in comparison with low, individuals reported sexual arousal and lust to be relatively high). Negative feelings were not associated with sexual desire, sexual behavior, or wanted sexual behavior. We found no support for a daily negative association between negative feelings and sexual expression.

[FIGURE 3 OMITTED]

Barlow (2002) offered a way to think about sexual occasions in which negative feelings incite sexual arousal. An excitation transfer process may occur when sexual arousal induced in association with anxiety (or other negative feelings) becomes incorporated into a response to sexual stimuli. Burleson and colleagues (2007) interpret their results with middle-aged women in a similar way when they state "if anxiety intensified arousal within each sexual episode, it might become associated with enhanced pleasure and hence increase the likelihood of sex occurring" (p. 365). In a sexual context high in sexual energy, anticipation of sexual involvement, and gratification, negative feelings are integrated into the excitement of sexual arousal and feelings of lust for one's partner. If negative feelings are at a level of intensity that is uncomfortable, but not at an intensity level that is preoccupying, then it is more likely to become incorporated into the excitement and gratification of sexual expression (Barlow, 2002; Kaplan, 1979). Participants in the current study reported relatively low levels of negative feelings (see Table 1), which may provide an emotional context for a positive association between negative feelings and sexuality, particularly sexual thoughts such as lust and sexual arousal. An alternative explanation for the positive association between negative feelings and sexual expression is that individuals may use sex to minimize negative emotions (Cooper, Shapiro, & Powers, 1998).

[FIGURE 4 OMITTED]

Positive associations were also found between negative feelings and both sexual arousal and sexual behavior when examined by couple type and gender. Positive associations were found for heterosexual women. On days when negative feelings toward one's partner were high versus days when they were low, sexual arousal and behavior were significantly higher. This same pattern was found for same-sex males. Heterosexual men showed a similar pattern for sexual behavior but not for arousal. No association was found between negative feelings and arousal or sexual behavior for lesbian individuals.

Based on results presented here, the positive association between negative feelings and the arousal or behavioral aspects of sexual expression exists for heterosexuals and gay males but not for women in lesbian relationships. Basson (2003) provided some clues for exploring why this may happen for women in heterosexual relationships. She argued that women become sexually aroused and sexually responsive in a different way than most men. As sexual arousal becomes elevated as a result of the male partners' sexual initiation, women become sexually interested and active, resulting in emotional and physical satisfaction. Emotional intimacy results from this process and becomes a part of the continued sexual response process. Thus, our results can be interpreted to indicate that women in heterosexual relationships are in an emotional and sexual environment that encourages excitation transfer processes to occur. An inspection of Figures 1 and 2 indicates that heterosexual women's sexual arousal and sexual behavior increase sharply as negative feelings increase. This supports Basson's (2003) observation that heterosexual women are in relationships with men, which may guide the movement toward greater sexual involvement that is consistent with their partners' higher level of sexual behavior. These increases create a sexually energized environment for women to move from lower levels of sexual interest to higher levels of sexual arousal and behavior and to incorporate negative feelings into the excitement of the sexual moments. This incorporation of negative feelings is consistent with the concept of excitation transfer. Gay men may be likely to jointly create a sexually energized relationship that activates an excitation transfer process, whereby negative feelings can be become a part of the sexual experience. The results for women in lesbian relationships reflect a less sexually charged environment that may discourage the excitation transfer process. Thus, negative feelings may not become incorporated into the sexual environment. When negative feelings become separated from sexual experiences, these experiences may lose potency to reduce or eliminate negative feelings.

Anger and sexual expression were not related in any significant way with the exception of a weak, positive association between anger and sexual desire. Perhaps the low level of daily anger was so low that it did not initiate an excitation process that could affect sexual expression. Anxiety and sadness, however, also were reported at relatively low levels, and both of these feelings were associated with both wanting and engaging in sexual behavior. In nondistressed relationships, participants might experience less anger and make additional efforts to separate these feelings from the appetitive needs expressed through sexuality (Christopher & Kisler, 2004).

The results for sadness and sexual expression varied by both gender and couple type. Perhaps most telling from this set of results is the relationship between sadness and sexuality for heterosexual women, because on days they are sad, sexual arousal, lust, sexual desire, and sexual behavior are high. In contrast, women in lesbian relationships report lower sexual arousal and lust on days they are sad. These results could be interpreted to suggest that the lower frequency of lust and sexual arousal in lesbian relationships might have something to do specifically with sadness suppressing sexuality, or that sexual expression is not sufficiently significant to reduce feelings of sadness. In contrast, heterosexual women, when sad, might allow a highly energized sexual context in which male partners experience elevated sexual arousal and behavior to convey a message that the partner cares about them and wants to help with reducing or eliminating the sadness (Coyne, 1990). Men in gay relationships are similar to heterosexual women in that they experience higher lust and engage in more sexual behaviors on days they are sad; however, sadness is unrelated to arousal.

Limitations and Future Directions

The current study had limitations that should be noted. Even though the strengths of this study were its daily assessments of feelings and sexuality and the inclusion of heterosexual, gay, and lesbian couples, they also posed difficulties. Our diary was 14 days, which is a relatively small window of recording daily experiences, and collecting data only one time per day may not represent the complete experiences of a full day. Further, given the restrictions of participants' available time, only a limited number of items could be assessed, which affects decisions about how to measure the complex constructs of emotional and sexual experiences. Specifically, the number of negative feelings that could be assessed was limited, leaving out many potentially negative emotional experiences. For example, anger, anxiety, and sadness clearly do not represent the full range of negative feelings that occur in close relationships. Additionally, despite the fact that participants were asked to report how they felt toward their partners, it is possible that the negative feelings measures that were chosen may have been capturing both a relational component and a general emotional state on each day. Next, participants were asked to complete the questionnaires independent from their partners, and time stamps were used to monitor this. Given the Internet-based design, however, it was not possible for the researchers to assure that partners' responses were completely independent (Ogolsky, Niehuis, & Ridley, in press). Additionally, the sampling procedure was not random, which limits generalizability to other populations. Further, because of the correlational design, causal inferences could not be made. Last, the current sample reported relatively low levels of negative feelings and relatively high levels of positive feelings. For this reason, the sizes of the effects of positive feelings were considerably larger than those of negative feelings. Future research should explore the association between negative feelings and sexual expression in a sample of chronically sad, anxious, or angry participants.

Despite these limitations, the current study has implications for research and practice. There is substantial utility in continuing to use a daily diary approach to address issues surrounding the complex interactions in dynamic systems such as close relationships. Although the data presented here help in understanding the role of negative feelings in sexual expression, future work needs to continue to explore the ebb and flow of positive and negative feelings and the full range of sexual experiences. Future research should examine the mechanisms that connect affect to sexual experiences. Additionally, despite our inclusion of individual differences (e.g., gender), the exploration of distal factors (e.g., ethnic and cultural factors, socioeconomic status) that influence feelings and sexuality should be explored. Professionals working with couples of various types should note the strong associations between positive feelings and sexual expression as well as the complex role that negative feelings play in sexuality. The results presented here clearly suggest that negative feelings do not necessarily suppress sexuality and in fact may enhance it, and that this association varies by gender and couple type. When individuals are up or down emotionally with their partners, sexual interaction must be negotiated. This negotiation becomes particularly challenging, however, when the emotional environment is negative and sex remains an expected and desired relational experience.

References

Aiken, L. S. & West, S. G. (1991). Multiple regression: Testing and interpreting interactions. Thousand Oaks, CA: Sage.

Araujo, A. B., Mohr, B. A., & McKinlay, J. B. (2004). Changes in sexual function in middle-aged and older men: Longitudinal data from the Massachusetts male aging study. American Geriatrics Society, 52, 1502-1509.

Bancroft, J., Janssen, E., Strong, D., & Vukadinovic, Z. (2003). The relation between mood and sexuality in gay men. Archives of Sexual Behavior, 32, 231-242.

Bancroft, J. & Vukadinovic, Z. (2004). Sexual addiction, sexual compulsivity, sexual impulsivity, or what? Towards a theoretical model. Journal of Sex Research, 41, 225-234.

Barlow, D. H. (2002). Anxiety and its disorders. New York: The Guilford Press.

Barlow, D. H., Sakheim, D. K., & Beck, J. G. (1983). Anxiety increases sexual arousal. Journal of Abnormal Psychology, 92, 49-54.

Basson, R. (2003). Women's difficulties with low sexual desire and sexual avoidance. In S. B. Levine (Ed.), Handbook of clinical sexuality for mental health professionals (pp. 111-130). New York: Brunner-Routledge.

Beck, A. T. (1967). Depression: Causes and treatment. Philadelphia: University of Pennsylvania Press.

Beggs, V. E., Calhoun, K. S., & Wolchik, S. A. (1987). Sexual anxiety and female sexual arousal: A comparison of arousal during sexual anxiety stimuli and sexual pleasure stimuli. Archives of Sexual Behavior, 16, 311-319.

Berscheid, E. (1988). Some comments on love's anatomy: Or, whatever happened to old-fashioned lust? In R. J. Sternberg & M. L. Barnes (Eds.), The psychology of love (pp. 359-374). New Haven, CT: Yale University Press.

Blumstein, P. & Schwartz, P. (1983). American couples." Money, work, sex. New York: Morrow.

Burleson, M. H., Trevathan, W. R., & Todd, M. (2007). In the mood for love or vice versa? Exploring the relations among sexual activity, physical affection, affect and stress in the daily lives of mid-aged women. Archives' of Sexual Behavior, 36, 357-368.

Cassidy, W. L., Flanagan, N. B., Spellman, M., & Cohen, M. E. (1957). Clinical observations in manic depressive disease. Journal of the American Medical Association, 164, 1535-1546.

Christopher, F. & Kisler, T. (2004). Sexual aggression in romantic relationships. In J. H. Harvey, A. Wenzel, & S. Sprecher (Eds.), The handbook of sexuality in close relationships (pp. 287-310). Mahwah, N J: Lawrence Erlbaum Associates.

Cooper, M. L., Shapiro, C. M., & Powers, A. M. (1998). Motivations for sex and risky sexual behavior among adolescents and young adults: A functional perspective. Journal of Personality and Social Psychology, 75, 1528-1558.

Coyne, J. C. (1990). Interpersonal processes in depression. In G. Keitner (Ed.), Depression and families (pp. 31-54). Washington, DC: American Psychiatric Press.

Dindia, K. (2000). Relational maintenance. In C. Hendrick & S. S. Hendrick (Eds.), Close relationships. A sourcebook (pp. 287-299). Thousand Oaks, CA: Sage.

Elliot, A. N. & O'Donohue, W. T. (1997). The effects of anxiety and distraction on sexual arousal in a nonclinical sample of heterosexual women. Archives of Sexual Behavior, 26, 607-624.

Fortenberry, J. D., Temkit, M., Tu, W., Graham, C. A., Katz, B. P., & Orr, D. P. (2005). Daily mood, partner support, sexual interest, and sexual activity among adolescent women. Health Psychology, 24, 252-257.

Giles, J. (2004). The nature of sexual desire. Westport, CT: Greenwood Publishing Group, Inc.

Huston, T. L. (2000). The social ecology of marriage and other intimate unions. Journal of Marriage and the Family, 62, 298-320.

Kaplan, H. (1979). Disorders of sexual desire: And other new concepts and techniques in sex therapy. New York: Brunner/Mazel.

Kelley, H. H. (1979). Personal relationships: Their structures and processes. Hillsdale, N J: Lawrence Erlbaum Associates, Inc.

Kennedy, S. H., Dickens, S. E., Eisfeld, B. S., & Bagby, R. M. (1999). Sexual dysfunction before antidepressant therapy in major depression. Journal of Affective Disorders, 56, 201-208.

Lang, P. J. (1994). The varieties of emotional experience: A meditation on James-Lange theory. Psychological Review, 110, 211-221.

Lykins, A., Janssen, E., & Graham, C. (2006). The relationship between negative mood and sexuality in heterosexual college women and men. The Journal of Sex Research, 43, 136-143.

Ogolsky, B. G., Niehuis, S., & Ridley, C. A. (in press). Using online methods and designs to conduct research on personal relationships. Marriage and Family Review.

Palace, E. M. & Gorzalka, B. B. (1990). The enhancing effects of anxiety on arousal in sexually dysfunctional and functional women. Journal of Abnormal Psychology, 99, 403-411.

Peplau, L. A., Fingerhut, A., & Beals, K. P. (2004). Sexuality in relationships of lesbians and gay men. In J. H. Harvey, A. Wenzel, & S. Sprecher (Eds.), The handbook of sexuality in close relationships (pp. 349-369). Mahwah, N J: Lawrence Erlbaum Associates.

Plant, E. A., Hyde, J. S., Keltner, D., & Devine, P. G. (2000). The gender stereotyping of emotions. Psychology of Women Quarterly, 24, 81-92.

Raudenbush, S. W. & Bryk, A. S. (2002). Hierarchical linear models: Applications and data analysis methods (2nd ed.). Thousand Oaks, CA: Sage.

Raudenbush, S., Bryk, A., Cheong, Y., & Congdon, R. (2001). HLM5: Hierarchical linear and nonlinear modeling. Chicago: Scientific Software International.

Regan, P. C. (2000). The role of sexual desire and sexual activity in dating relationships. Social Behavior and Personality, 28, 51-60.

Regan, P. C. & Berscheid, E. (1999). Lust: What we know about human sexual desire. Thousand Oaks, CA: Sage.

Ridley, C. A., Cate, R. M., Collins, D. M., Reesing, A. L., Lucero, A. A., Gilson, M. S., & Almeida, D. M. (2006). The ebb and flow of marital lust: A relational approach. The Journal of Sex Research, 43, 144-153.

Rusbult, C. E. (1983). A longitudinal test of the investment model: The development (and deterioration) of satisfaction and commitment in heterosexual involvements. Journal of Personality and Social Psychology, 45, 101-117.

Russell, J. A. (1980). The circumplex model of affect. Journal of Personality and Social Psychology, 39, 1161-1178.

Russell, J. A. & Carroll, J. M. (1999). On the bipolarity of positive and negative affect. Psychology Bulletin, 125, 3-30.

Vohs, K. D., Catanese, K. R., & Baumeister, R. F. (2004). Sex in "his" versus "her" relationships. In J. H. Harvey, A. Wenzel, & S. Sprecher (Eds.), The handbook of sexuality in close relationships (pp. 554-74). Mahwah, N J: Lawrence Erlbaum Associates.

Wenzel, A., Jackson, L. C., & Brendle, J. R. (2004). Psychopathology, sexuality and the partner relationship. In J. H. Harvey, A. Wenzel, & S. Sprecher (Eds.), The handbook of sexuality in close relationships (pp. 545-572). Mahwah, N J: Lawrence Erlbaum Associates.

Carl Ridley

The University of Arizona 

Brian Ogolsky

The University of Texas at Austin 

Pamela Payne, Casey Totenhagen, and Rodney Cate

The University of Arizona

Correspondence should be addressed to Carl Ridley, The University of Arizona, Family and Consumer Sciences Building, 650 N Park Ave., Tuscon, AZ 85721-0078. E-mail: [e-mail]

(1) Sexual behaviors were first tested individually: however, because no significant differences emerged between the sexual behaviors, they were summed to create a more parsimonious scale.

Table 1. Means and Standard Deviations of Study Variables
for Heterosexual, Gay, and Lesbian Couples

                                      Gay        Lesbians
Variables         Heterosexuals      Males

Negative Affect    0.51 (0.74)    0.49 (0.73)   0.37 (0.63)
Positive Affect    3.08 (0.87)    3.01 (0.86)   3.27 (0.79)
  Anger            0.41 (0.82)    0.36 (0.78)   0.32 (0.76)
Sadness            0.51 (0.90)    0.46 (0.86)   0.41 (0.82)
  Anxiety          0.62 (0.95)    0.65 (0.97)   0.38 (0.77)
Sexual Arousal     2.37 (1.32)    2.24 (l.27)   2.16 (l.27)
Lust               1.68 (1.39)    1.67 (1.35)   1.55 (1.31)
Sexual Desire      2.36 (l.34)    2.26 (l.29)   2.19 (l.26)
Wanted Sexual      2.81 (1.26)    2.53 (1.31)   2.43 (1.37)
  Behavior
Sexual Behavior    2.61 (1.39)    0.85 (0.87)   0.64 (0.76)

Table 2. Multilevel Model of the Effects of Negativity, Positivity,
Gender, and Couple Type on Sexual Desire, Sexual Arousal, Lust,
Wanted Sexual Behavior, and Sexual Behavior

                                         Sexual
Fixed Effects                            Arousal           Lust

Intercept                             -.81 (.58)       -1.40 (.60) *
Negative Feelings                      .30 (.11) **      .32 (.11) *
Positive Feelings                      .92 (.17) ***     .70 (.19) ***
Gender                                -.28 (.11) **     -.09 (.12)
Couple Type                           -.20 (.58)        -.52 (.61)
Negative Feelings x Gender            -.19 (.15)        -.22 (.13)
Negative Feelings x Couple
  Type                                 .06 (.12)         .03 (.11)
Positive Feelings x Gender            -.03 (.03)        -.01 (.04)
Positive Feelings x Couple
  Type                                 .14 (.17)        -.15 (.18)
Gender x Couple Type                  -.23 (.11) *      -.18 (.12)
Negative Feelings x Gender x
  Couple Type                          .07 (.03) *       .05 (.03)
Positive Feelings x Gender x
  Couple Type                          .06 (.03) *       .05 (.03)

                                        Sexual        Wanted Sexual
Fixed Effects                           Desire         Behavior

Intercept                            -1.23 (.59) *      0.76 (.93) *
Negative Feelings                      .21 (.15)        -.05 (.21)
Positive Feelings                     1.03 (.16) ***     .58 (.22) *
Gender                                -.27 (.11) *      -.05 (.92)
Couple Type                           -.65 (.68)        -.53 (.93) *
Negative Feelings x Gender             .01 (.02)        -.00 (.03)
Negative Feelings x Couple
  Type                                -.18 (.14)         .18 (.17)
Positive Feelings x Gender             .03 (.03)        -.01 (.04)
Positive Feelings x Couple
  Type                                 .08 (.19)        -.04 (.20) *
Gender x Couple Type                  -.19 (.11)        -.17 (.53)
Negative Feelings x Gender x
  Couple Type                          .04 (.03)         .02 (.03)
Positive Feelings x Gender x
  Couple Type                          .06 (.03) *       .08 (.14)

                                        Sexual
Fixed Effects                          Behavior

Intercept                              .82 (1.03)
Negative Feelings                      .10 (.22)
Positive Feelings                      .50 (.28) ***
Gender                                 .06 (.12)
Couple Type                           1.68 (1.03)
Negative Feelings x Gender            -.00 (.03)
Negative Feelings x Couple
  Type                                -.00 (.16)
Positive Feelings x Gender            -.04 (.03)
Positive Feelings x Couple
  Type                                -.47 (.28)
Gender x Couple Type                  -.40 (.13) **
Negative Feelings x Gender x
  Couple Type                          .08 (.03) **
Positive Feelings x Gender x
  Couple Type                          .13 (.03) **

Note. Coefficients are reported as beta coefficients with the
accompanying standard error following in parentheses.

*** p < .001. ** p <.01. * p < .05.

Table 3. Multilevel Models of the Effects of Anger, Anxiety, Sadness,
Positivity, Gender, and Couple Type on Sexual Desire, Sexual
Arousal, Lust, Wanted Sexual Behavior, and Sexual Behavior

                              Sexual
Fixed Effects                 Arousal           Lust

Intercept                    -.76 (.60)     -1.32 (.64) *
Anger                         -20 (.12)       .12 (.11)
Anxiety                       .18 (.12)       .19 (.13)
Sadness                      -.07 (.14)       .01 (.15)
Positive Feelings             .90 (.16) ***   .68 (.19) **
Gender                       -.29 (.11) **   -.10 (.12)
Couple Type                  -.16 (.60)      -.60 (.64)
Anger x Gender               -.01 (.02)       -03 (.02)
Anger x Couple Type           .01 (.12)       .12 (.11)
Anxiety x Gender              .00 (.02)       .01 (.02)
Anxiety x Couple Type         .10 (.12)      -.03 (.13)
Sadness x Gender              .02 (.02)      -.02 (.02)
Sadness x Couple Type        -.12 (.14)      -.03 (.15)
Positive Feelings
  x Gender                    .03 (.03)      -.01 (.04)
Positive Feelings x
  Couple Type                -.15 (.17)      -.14 (.19)
Gender x Couple Type         -.23 (.11) *    -.18 (.12)
Anger x Gender x
  Couple Type                -.01 (.02)      -.01 (.02)
Anxiety x Gender x
  Couple Type                 .00 (.02)      -.01 (.02)
Sadness x Gender x
  Couple Type                 .07 (.02) **    .07 (.02) **
Positive Feelings x
  Gender x Couple Type        .06 (.03) *     .05 (.04)

                              Sexual        Wanted Sexual
Fixed Effects                 Desire          Behavior

Intercept                   -1.19 (.62) *     .90 (.98)
Anger                         .22 (.11) *    -.01 (.17)
Anxiety                       .06 (.12)       .48 (.16) **
Sadness                      -.07 (.19)      -.38 (.15) *
Positive Feelings            1.02 (.16) ***   .50 (.23) *
Gender                       -.30 (.11) **    .02 (.14)
Couple Type                  -.66 (.62)      -.22 (.98)
Anger x Gender               -.01 (.02)       .03 (.03)
Anger x Couple Type           .07 (.11)       .06 (.17)
Anxiety x Gender              .00 (.02)       .02 (.03)
Anxiety x Couple Type        -.00 (.12)       .56 (.16) **
Sadness x Gender              .03 (.02)      -.06 (.03) *
Sadness x Couple Type        -.17 (.19)      -.47 (.15) **
Positive Feelings
  x Gender                    .03 (.03)      -.02 (.04)
Positive Feelings x
  Couple Type                -.03 (.16)      -.16 (.23)
Gender x Couple Type         -.20 (.11)      -.24 (.14)
Anger x Gender x
  Couple Type                -.01 (.02)       .01 (.03)
Anxiety x Gender x
  Couple Type                -.00 (.02)      -.03 (.03)
Sadness x Gender x
  Couple Type                 .05 (.02) *     .04 (.03)
Positive Feelings x
  Gender x Couple Type        .06 (.03) *     .08 (.04) *

                              Sexual
Fixed Effects                Behavior

Intercept                    .79 (1.06)
Anger                         .18 (.21)
Anxiety                       .35 (.16) *
Sadness                      -.39 (.27) *
Positive Feelings             .50 (.29) *
Gender                        .07 (.13)
Couple Type                  1.59 (l.06)
Anger x Gender                .00 (.03)
Anger x Couple Type           .06 (.21)
Anxiety x Gender             -.02 (.02)
Anxiety x Couple Type         .27 (.16)
Sadness x Gender              .01 (.02)
Sadness x Couple Type        -.41 (.27)
Positive Feelings
  x Gender                   -.04 (.03)
Positive Feelings x
  Couple Type                -.46 (.29)
Gender x Couple Type         -.43 (.13) **
Anger x Gender x
  Couple Type                 .05 (.03)
Anxiety x Gender x
  Couple Type                -.01 (.02)
Sadness x Gender x
  Couple Type                 .05 (.02) *
Positive Feelings x
  Gender x Couple Type        .13 (.03) ***

Note. Coefficients are reported as beta coefficients with the
accompanying standard error following in parentheses.

*** p < .001. ** p < .01. * p < .05.